GBDT
GBDT copied to clipboard
为什么没有subsample
我记得GBDT里有shrinkage和subsample的过程啊 代码里只有列采样
代码里有shrinkage与subsample这个啊,如下: shrinkage: for node in leaf_nodes: for id in node.get_idset(): f[id][label] += learn_rate*node.get_predict_value() subsample: subset = sample(subset, int(len(subset)*self.sample_rate)) 还是说我理解错了?
@zhaochangming 你是对的